Industry 4.0 is the most recent industrial revolution that aims to improve not only the productivity in the 21 st century, but also the flexibility, adaptability, and resilience of the industrial systems. It enables the collection of real-time data from industrial systems, Thanks to the development of Internet of Things (IoT) technology. Hence, analyzing online collected data enables to deal with several industrial issues in real-time such as machines' break-or slowdowns , quality crisis, flows disruptions, etc. In traditional industrial systems, previous works focused on both scheduling and rescheduling schemes in order to improve the system performance. However, few works dealt with system disruption monitoring due to the lack of real-time data about the system running. Furthermore, the remote and constant monitoring amenities were not established yet, properly. In this paper, we propose a system disruption monitoring tool in Industry 4.0 system. Our tool focuses on system disruption related to resources localization, or when a resource is in an unexpected location. Thus, a machine learning algorithm is used to generate a prediction model of resources localization by considering the real tasks scheduling in terms of resources localization. Therefore, as real resources localization can be collected from the industrial system through the IoT network, our tool enables to detect system disruption, risk, in real-time when comparing predicted localization to the real one. Moreover, our tool is executed in a Fog computing architecture which is emerging as an extension of cloud computing to provide local processing support with good latency. The experimental results show the efficiency of our tool in terms of prediction accuracy and time complexity when compared to other machine learning algorithms, in addition to its ability to control and detect system disruption in real-time.